Maltreated foster children have a high prevalence of developmental delays, psychosocial maladjustment, and psychiatric symptoms (Klee, Kronstadt, & Zlotnick, 1997; Landsverk, Garland, & Leslie, 2002). Recent work has focused on understanding effects of maltreatment and caregiver transitions on neurobiological systems involved in the stress response, specifically the hypothalamic-pituitary-adrenal (HPA) axis (Bruce, Fisher, Pears, & Levine, 2009; Dozier et al., 2006). The HPA axis—indexed by cortisol levels—displays a diurnal rhythm, typically characterized by relatively high morning levels that decline to near-zero by bedtime (Kirschbaum et al., 1990). This change across the day is referred to as the diurnal cortisol slope. Previous work indicates that young foster children are more likely than their nonmaltreated peers to evidence a flattened diurnal cortisol slope (or less change across the day), resulting from low morning cortisol levels, which may indicate a poorly regulated system (Dozier et al., 2006; Fisher et al., 2006). However, despite the high prevalence of socioemotional difficulties in this population, only a minority of foster children display low morning cortisol levels (Bruce et al., 2009). Because the HPA axis is also designed to respond to physical and psychological challenges in the environment (Kirschbaum et al., 1990), examining the responsiveness of the system to stressors may provide further insight into potential effects of maltreatment and caregiver transitions. Although research with older children and adults has often employed laboratory stressors to measure immediate changes in cortisol levels, research with normative samples of young children has focused on changes in the diurnal cortisol rhythm in response to naturally occurring stressors, such as the start of school (Bruce, Davis, & Gunnar, 2002; Davis, Donzella, Krueger, & Gunnar, 1999; Quas, Murowchick, Bensadoun, & Boyce, 2002; Turner-Cobb, Rixon, & Jessop, 2008). This work provides evidence for a transient perturbation in the diurnal cortisol rhythm in response to the start of school, characterized by a steeper slope on the 1st day of school compared to weekend days later in the year (Bruce et al., 2002). This steeper slope, which has been viewed as a healthy adaptation to a novel, significant event, resulted from higher morning and lower evening cortisol levels on the 1st day of school. However, individual differences appear to play a large role, such that higher temperamental surgency (Bruce et al., 2002; Davis et al., 1999) and lower effortful control (Turner-Cobb et al., 2008) predict a steeper slope beyond the 1st day of school, potentially indicative of greater reactivity or less recovery of the system. Although Gutteling, de Weerth, and Buitelaar (2008) documented effects of prenatal stress on the diurnal cortisol rhythm across the start of school, the effects of severe early life stress, such as maltreatment and caregiver transitions, remain unknown. Moreover, it is unclear whether comprehensive early preventive intervention programs such as Multidimensional Treatment Foster Care for Preschoolers (MTFC-P; Fisher, Ellis, & Chamberlain, 1999) affect changes in the diurnal cortisol rhythm in response to a naturally occurring stressor. MTFC-P supports foster parents in providing consistent and contingent responses to positive and negative child behaviors. The child’s environment thus becomes more predictable, responsive and reinforcing, which has been shown to prevent the development of an increasingly flattened diurnal cortisol rhythm in foster children (Fisher, Stoolmiller, Gunnar, & Burraston, 2007). However, the impact of MTFC-P on the responsiveness of the HPA axis to important life events, such as the start of school, remains untested. Foster children often experience difficulties in school (Pears, Fisher, Bruce, Kim & Yoerger, 2010), and preventive interventions that influence the capacity of the HPA axis to respond to the challenge of the start of school may support a more adaptive transition. This pilot study examines changes in diurnal cortisol in response to the start of school for three groups of children: foster children who received regular foster care (RFC); foster children who received MTFC-P; and low-income, nonmaltreated children living with their biological parents (community comparison [CC]). We hypothesized that RFC children would show an extended cortisol response, characterized by a steeper slope on the 1st and 5th days compared to the week before school, and that MTFC-P and CC children would show a transient cortisol response, characterized by a steeper slope only on the 1st day of school.
Read full abstract